Summary African Americans are at significantly greater risk of developing type 2 diabetes than non-Hispanic whites. Despite the long-standing recognition of this racial disparity and attempts to eliminate it, it persists. Diabetes is associaed with reduced quality of life, decreased life expectancy and increased economic burden, and therefore African Americans experience a greater health burden than non-Hispanic whites. Thus, more efforts to combat this disparity in diabetes is urgently needed in order to achieve health equity for African Americans. Laboratory and observational studies have observed (1) significant associations between deficient sleep and increased diabetes risk and (2) shorter sleep durations and poorer sleep quality in African Americans compared to whites. No research to date, however, has explicitly examined whether deficient sleep partially accounts for the increased diabetes risk in African Americans. Therefore, we propose to examine sleep duration, sleep quality and circadian disruption as novel biobehavioral mediators of the racial disparity in diabetes risk. Indeed, despite it being well-recognized that (1) short sleep durations alter circadian phase, (2) circadian disruption produces metabolic disturbances, and (3) the melatonin rhythm directly impacts beta-cell function, circadian disruption has not previously been considered as a key mediator between sleep and diabetes risk. Thus the 3 primary goals of this study are to (1) conduct comprehensive and ecologically-valid assessments of sleep in people's homes, (2) assess circadian phase in the home environment by objectively controlling for light exposure and timing of sample collection and (3) determine if differences in home sleep and circadian measures partially explain differences in diabetes risk factors between African Americans and non- Hispanic whites. Our primary marker of diabetes risk will be the Matsuda Index of insulin sensitivity, estimated from a frequently-sampled oral glucose tolerance test. Secondary outcomes will include area-under-the-curve of glucose levels, disposition index, and a marker of inflammation (C - reactive protein), blood pressure, heart-rate variability, and lipid profiles. We plan to study 72 African Americans and 72 non-Hispanic whites without diabetes or sleep disorders. We have preliminary data that 1) support the existence of racial differences in the duration, structure and timing of sleep, (2) support our hypothesis that deficient sleep may mediate some health disparities, (3) support the existence of racial differences in circadian timing, (4) demonstrate the feasibility of conducting unattended in-home PSG, (5) demonstrate the feasibility and validity of data from at- home circadian phase assessments and (6) demonstrate the feasibility of the complete shortened protocol to be used in this study. This important multidisciplinary project will yield unique insights into the bio behavioral mechanisms underlying increased diabetes risk in African-Americans and will likely lead to the development of novel targeted interventions for diabetes including bedtime extension or regularity and chrono-therapeutic treatments. This application is in response to PA-09-262 Health Disparities in NIDDK Diseases.